Project Summary: Despite extraordinary advances in potent antiretroviral therapies that are capable of altering the course of the disease for many patients, HIV research continues to present new challenges whose study and resolution require innovative new statistical methodology for design and analysis of HIV clinical trials. In the extension period of this MERIT award, new statistical methods will be developed that offer cutting-edge solutions to fundamental and emerging challenges in HIV clinical research that may be translated into accessible tools for data analysts. Lifelong commitment to antiretroviral therapy is complicated by side-effects, toxicities, drug resistance, costs, and life style factors, and developing strategies for use of these therapies over time that can circumvent these issues and still sustain viral suppression and preserve immunological function are a central focus of HIV research. In the first project, a realistic, principled framework for trial design and analysis in which such time-dependent HIV treatment strategies may be conceived and evaluated feasibly will be developed. Clinical trials focusing on the difference in two treatments are a mainstay of HIV research, but the analysis of such trials is often complicated by missing data and uncertainty over whether and how to incorporate additional auxiliary information on the subjects. In the second project, a unified statistical framework to address these issues that will lead to accessible techniques for practitioners will be developed. Characterizing the relationship between longitudinal measures of biomarkers (e.g., CD4, viral load) and long-term clinical endpoints such as disease progression is of widespread interest in HIV research;e.g., models for this purpose are an important tool in the evaluation of biomarkers as potential surrogate endpoints. In the third project, a new, computationally feasible approach that leads to efficient inferences will be formulated. Relevance: This project will provide HIV researchers with new tools to conceive and evaluate new treatment strategies for HIV infection, including strategies for how best to treat HIV-infected individuals over time while avoiding problems of drug resistance and side effects. New methods for data analysis to help researchers understand the relationships between markers such as CD4 and disease progression will also be developed.